Extensive studies have revealed that in contrast to normal tissue, prostate carcinomas (PCa) and other high mortality cancers including glioblastomas, and small cell lung carcinomas are critically dependent on the uptake of the non-essential amino acid L-cysteine (L-Cys) and its oxidized form L-cystine (CSSC) for proliferation and survival (due to increased requirements for anti-oxidants and in some tumors because they display defects in endogenous L-Cys synthesis). We hypothesize that enzyme-mediated systemic depletion of the serum L-Cys/CSSC pool (using an engineered human enzyme, and thus likely well tolerated) constitutes a powerful and completely novel cancer therapeutic approach for PCa as a monotherapy and as an agent that is mechanistically synergistic with existing chemotherapeutic modalities. This hypothesis is supported by strong preliminary data demonstrating that the administration (every 4 days) of a prototypical engineered human L- Cys/CSSC depleting enzyme in an allograft model of PCa: (i) results in near complete depletion of the serum L-Cys/CSSC pool; (ii) mediates sustained and complete cessation of tumor growth with cancer cells experiencing cell cycle arrest accompanied by increased oxidative stress and autophagy and (iii) importantly treatment for over a month was very well tolerated with no weight loss or gross toxicities. To build on these preliminary observations we present a detailed research plan that is aimed at delivering a clinically relevant enzyme therapeutic candidate and providing the validation by in vitro and in allograft/xenograft PCa models that will eventually hel support late stage preclinical development and an IND application. While L-Cys/CSSC depletion may be applicable to the treatment of many tumors we elected to focus on PCa first because many human PCA tumors are known to experience defects in L-Cys/CSSC homeostasis and second because effective non-toxic treatments are needed as PCa is the second leading cause of cancer death in American men. Reaching these goals will require the completion of work under 3 Specific Aims: Sp. Aim 1 will focus on the engineering of a candidate human enzyme for L-Cys/CSSC depletion that displays optimal enzyme kinetics, substrate selectivity, stability to deactivation and aggregation and facile production by bioprocessing. The PK and PD of the optimized enzymes as well as any toxic effects upon repeated administration will be evaluated in detail. In Sp. Aim 2 we will seek to understand the global mechanistic/metabolic consequences of L-Cys/CSSC depletion on PCa cells, what biomarkers may predict susceptibility/resistance and to determine which combinations with standard of care chemotherapeutics may provide synergistic/additive effects. Finally, in Sp. Aim 3 we will evaluate the optimized enzyme in clinically relevant orthotopic PCa animal models that represent diverse phenotypes (i.e. p53 status, androgen dependence/independence) and mouse orthotopic PCa allografts such that the efficacy of the drug can be studied in models closely mimicking the actual disease.